Infection in rabbits injected with HIV-1 has been detected by seroconversion, detection of virus in cells and organs using PCR and in situ hybridization, and by isolation of virus from PBMC. While the fact that rabbits can be infected with HIV is promising, the slow course of infection, difficulty in isolation of virus and the absence of consistent disease signs in infected rabbits have limited utility of this model for AIDS. Information gained in recent efforts aimed at improving the model suggest that infection in the rabbit may be enhanced in several ways. Adaptation of HIV to grow more efficiently in rabbits has been attempted by long term in vivo passage; injection of as little as 1 ml of blood from animals receiving splenocytes from a rabbit infected with HIV-1 and held for 2.3 years has resulted in lymphopenia, CD4 T cell depletion and for 2 of 12 rabbits, death. Initial results from structural analyses of HIV-1 amplified from a spleen cell recipient shows nucleotide sequence identity of 85% to the original input isolate, HIV/Lai. In order to determine whether the slow progression of infection in rabbits is rooted in problems with transcription of viral DNA has prompted preparation of CAT constructs utilizing the 5' or 3' LTRS from HIV-1 to provide the promoter or 3' processing signals. These are being used to compare LTR-mediated gene expression in rabbit and human cells. There are no apparent defects in promoter or processing of genes flanked by HIV-1 LTR's. In subsequent experiments, molecular clones of HIV-1 were used in transfection experiments to assess production of infectious virus in rabbit as compared to human cell lines. Both species produced virus that appeared normal by all criteria, including ability to infect human cells. The rabbit lines produced a somewhat higher ratio of defective particles.